A. Field of the Invention
The present invention is related to imaging methods and apparatuses and more particularly to a method and apparatus for obtaining 2D silhouettes from one or more 3D objects.
B. Description of the Related Art
Obtaining accurate 2D contours of a 3D object is traditionally accomplished via a traditional (focusing camera) optical system, these image acquisition systems employ a lens that focuses light at an imaging sensor allowing for an image of an object larger than the lens or the sensor, to be captured in a single action, without the need to move the camera system.
This process is not optimal for producing a real 2d representation of an object, because it introduces projection errors that distort the dimensions of the objects farther or closer to the lens. This is to say, that an object closer to the lens will be shown larger than an identical object placed farther away from the lens. Although this can be compensated in flat objects by placing a reference in the same plane of capture, a 3D object does not have a single plane of capture hence complicating this kind of approach (a cone with its point toward the lens can actually be seen as a tube if it is close enough to the lens).
Furthermore, since the point of capture can be occluded by the same object being captured, a bigger protrusion on the back of the object can be covered by a smaller one closer to the lens.
This usually requires a great deal of manual labor after the capture is done and requires access to the object being captured, negating some of the advantages of an automated capture process.
There are special lenses that capture only the light travelling perpendicular to the lens, this allows for an accurate representation of the 3d object but can only accurately capture an object that is slightly smaller than the actual lens size. Because of this limitation and the size of theft construction relative to the capture size, these lenses are not practical for medium to large scale objects, and are only useful to inspect smaller objects like microelectronics or small screws and pipes.
In view of the above referred problems, applicant developed a method and apparatus for obtaining 2D silhouettes from one or more 3D objects, which uses an image acquisition unit having a telecentric lens system that is able to fully scan an object in one dimension.
The lens used in the method and apparatus of the present invention, only captures light travelling perpendicular to its surface and therefore, it doesn't suffer the projection distortion commonly associated with regular imaging systems.